Vmware, Inc.

Chen Doar

Rami Stern

Or Lupovitz

Meytal Genah

Dmitry Rabinovich

Shadi Massalha

Ehud Sicron

Dima Golbert

Boaz Harel

Anton Gartsbein

Igal Bakshan

Lior Hipsh

A system is described for downloading server-based content to peers in a P2P network. The server-based content can be partitioned and different peers can be given priority for downloading the different parts. A client on each peer can view what content is available on each of the other peers as well as what content is available on a virtual peer, which represents content on the server and only makes content that exists on the server and not on any peers available for download. Consequently, a client on a peer requesting to download the content can download the content from the peers and the virtual peer based on a file sharing protocol, such as BitTorrent.

Downloading of server-based content through peer-to-peer networks

Techniques are described for delivering an application to a computing device by dynamically creating an application storage volume, such as a virtual disk, based on a minimal subset of application files downloaded from a remote server and attaching the virtual disk to the computing device in order to make the application executable from the virtual disk. The minimal subset of files access to at least some portion of the functionality of the application. Dynamically creating the virtual disk can be completed quickly and the minimal subset of application files is smaller than the full size of the application, which improves the speed of application delivery to the device. After the virtual disk is attached, the user can begin working on the application, while the remaining files are streamed to the virtual disk both as an ongoing background process and in response to on-demand requests, transparently from the user's perspective.

Application delivery to a computing device using dynamic application volume creation and streaming

Described embodiments provide ways to optimize application minimal sets based on the behavior of users. Described embodiments further provide ways to produce a minimal set that is optimized based on various attributes regarding the target endpoint onto which the minimal set is to be deployed, such as the target endpoint's operating system, applications installed on the target endpoint, hardware on the target endpoint, or profile information regarding the target endpoint's user. Also, described embodiments provide ways to prioritize the order of deployment when multiple minimal sets for various applications are downloaded to a target endpoint.

Using machine learning to optimize minimal sets of an application

Distributing an update from a server to multiple endpoint devices that reside on the same local network. The server provides information to each client device identifying all other devices located on the same local network that will also need to receive the update. Each client device sorts the files and begins the process of downloading the files by checking, for each file in sequence, whether another device on the same network has downloaded the file or is currently in the process of downloading. If another device on the same network has completed downloading the file, the device requests the file from the other client device. If another device is currently in the process of downloading, the device places the file at the end of the sequence. If the file has not been downloaded and is not being downloaded, the endpoint device initiates downloading the file from the server.

Peer-to-peer network download optimization

Local distributed backup techniques are described, where each client device calculates a manifest of its files and sends the manifest to a central server. The server performs deduplication based on the received manifest to determine which files are missing and need to be backed up, and notifies the client of those files. At this point, instead of uploading the missing files to the central server, the files are instead uploaded to another peer client device that has sufficient disk space and resides in the network vicinity of the original client device. The upload may be done to multiple peers for redundancy. Upon backup completion, the client state may be left as-is, in which case the client periodically uploads incremental updates to keep any changes synchronized with the backup peer. Alternatively, the client can be centralized to the central server, but at a later time when more resources are available.

Growing community of inventors

Herzliya, Israel

Chen Doar